3.1 Characterization of the BGPs
3.1.1 Morphology
The optical microscope and SEM photographs of BGPs are shown in Fig. 2(A) and Fig. 2(B). The bioactive glass particles were spherical. The inside of the spheroid was hollow. Fig. 2(C) shows the particle size distribution of BGPs. The particle size of BGPs mainly ranged from 15 to 55 μm. BGPs with particle sizes between 30 and 50 μm accounted for more than 75% of the total BGPs. The particle size follows a normal distribution. This indicated that the BGPs had good homogeneity.
3.1.2 FT-IR Spectra
The FT-IR spectra of the BGPs are shown in Fig. 2(D). The peak near 1090 nm is the stretching vibrational peak of Si-O. The peaks near 465 nm and 800 nm are the bending vibrational peak of Si-O-Si and the vibration peak of silica tetrahedral, respectively. The peaks near 560 nm and 605 nm are the bending vibrational peaks of P-O. All peaks are the characteristic peaks of the infrared spectrum of bioactive glass.
3.1.3 BSA Loading and Release
As shown in Fig. 2(E), the adsorption quantity of BSA to BG and BGP increased gradually over time. After 30 min, the protein adsorption rate of BGP group peaked the value of 88.2%, while the maximum protein adsorption rate of BG was 65.1% at 45 min. The protein adsorption rate of BGP was higher than that of BG. The BSA release is shown in Fig. 2(F). A protein burst occurred in the BG group. The release rate first rapidly rose to 58% in 10 mins, then fell back to approximately 45% and remained constant. In contrast, the release rate of the BGP group gradually increased to 33% within 20 minutes, then dropped below 30% and remained stable. BGP has advantages over BG in controlling protein release.
3.2 Characterization of the composites
3.2.1 Effect of Different Nef Contents on MC-3T3-E1 Cells
To determine the applied dosage of Nef, the proliferation and alkaline phosphatase (ALP) activity of MC-3T3-E1 cells planted on composite membranes containing different Nef contents were detected. According to the results in Fig. 3(A-C), Nef at low concentrations (1.56 μM~12.5 μM) had no obvious toxicity to cells, while Nef at concentrations over 25 μM significantly inhibited cell proliferation (p<0.05). However, at day 7 of cell culture, the inhibitory effect of the 25 μM and 50 μM treatments on cell proliferation decreased, and there was no significant difference between the 25 μM and control groups (p>0.05).
The results of ALP activity are shown in Fig. 3(D-E). Composites of PLGA/BGP/25N could promote the of MC-T3-E1 cells on day 3 (p<0.05). All composites containing Nef could promote the of MC-T3-E1 cells on day 7 (p<0.05). However, the ALP activity of the PLGA/BGP/25N and PLGA/BGP/50N groups was the highest and significantly higher than other groups (p<0.05). Based on the results of proliferation and ALP activity assays, Nef concentrations of 25 μM and 50 μM were selected for further research in this study.
3.2.2 Morphology
The microscopic structures observed by ESEM and micro-CT are shown in Fig. 4(A-E). The whole composite material presents a loose and porous structure. The porosity is up to 86.35%. The pore structure of the central part is different from that of the peripheral part. The center part presents a homogeneous honeycomb structure. Radial vascular channels with pipe diameters ranging from 100 μm to 150 μm exist around the periphery, extending from the center to the edge of the composite. A large amount of BGP is evenly distributed in the composite.
3.2.3 FT-IR Spectra
The FT-IR spectra of the composites are shown in Fig.4(F). In addition to the characteristic peaks of the infrared spectrum of bioactive glass, the characteristic peak of the phenolic hydroxyl group can be seen near 3280 nm, which indicated the existence of Nef.
3.2.4 In vitro Nef Release Profile
The release behavior of Nef in 24 hours is shown in Fig.4(G). It exhibited an initial burst release of Nef within the first half hour, which was determined to be 18.7% in the PLGA/BGP/25N group and 12.2% in the PLGA/BGP/50N group. Then, the drug showed sustained release, and the cumulative release of Nef reached 23.4% and 14.9% and was still continuing to be released slowly.
3.2.5 Anti-osteoclastogenic Effect of Composites
The effects of the composites on osteoclast differentiation were examined. M‐CSF‐dependent macrophages were cotreated with RANKL and cultured on different composites until large multinucleated “pancake”‐shaped osteoclasts were observed in the RANKL‐treated controls. The fixed cells were further stained for TRAP activity. As demonstrated in Fig. 5, Nef inhibited the total number of TRAP‐positive osteoclasts formed in response to RANKL. The TRAP activity was further detected by Tartrate Resistant Acid Phosphatase Assay Kit (Beyotime, Beijing). According to result of TRAP activity assay, the TRAP activity in the Nef groups was significantly higher than that in the control and PLGA/BGP groups (p<0.05). The results show that the composites containing Nef can inhibit RANKL‐induced osteoclast formation.
3.2.6 Gene Expression Analysis via qRT‒PCR
To explore the effect of different composites on the osteogenic differentiation of MC-3T3-E1 cells at the molecular level, the real-time quantitative PCR (qRT-PCR) was used to quantitatively measure the osteogenic differentiation-related target genes Runt-related transcription factor-2 (RUNX2), type I collagen (Col1), Osteopontin (OPN), and Osteocalcin (OCN). As shown in Fig. 6, MC-3T3-E1 cells were cocultured with different composites for 3 days, 7 days and 14 days. For Col1, there was no significant difference between each group at day 3. After 7 days of culture, compared with the PLGA group, the expression levels of Col1 genes were increased significantly in the PLGA/BGP and PLGA/BGP/25N groups. The expression levels in PLGA/BGP/50N were similar to those in the control group. In addition, with the increase in Nef content, the expression level of Col1 decreased gradually.
The composites containing Nef promoted the expression levels of RUNX2 at day 3. However, on day 7, RUNX2 expression was significantly higher in the PLGA group than in the other groups. However, RUNX2 expression in the PLGA/BGP/50N group was still higher than that in the PLGA/BGP group.
Gene expression of OPN was measured at days 3, 7 and 14. There was no significant difference between the groups on day 3. After 7 days of culture, compared with the PLGA group, the expression levels of OPN genes were increased significantly in the PLGA/BGP and PLGA/BGP/25N groups. The expression levels of OPN in the PLGA/BGP/50N group were also higher than those in the PLGA group. However, the difference was not statistically significant. The results of day 14 showed that the expression levels of OPN in the PLGA/BGP/25N and PLGA/BGP/50N groups were obviously promoted when compared to the other two groups. The expression level of OPN increased gradually with increasing Nef content.
Gene expression of OCN was measured at days 7 and 14. At 7 days, the expression of OCN in the PLGA/BGP, PLGA/BGP/25N and PLGA/BGP/50N groups was obviously upregulated compared with that in the PLGA group. The expression level of PLGA/BGP/25N was the highest. At 14 days, the OCN expression in the PLGA/BGP/25N group was still highest. The expression levels of OCN in the PLGA/BGP/25N and PLGA/BGP/50N groups were obviously upregulated compared to those in the other two groups.
IGF-1 plays an important role in bone repair. IGF-1R is a key receptor for IGF-1. In this study, the gene expression of IGF-1 and IGF-1R in MC-3T3-E1 cells was detected at 3 and 7 days. The IGF-1 expression level in the PLGA group was significantly lower than that in the other three groups at each timing. However, at day 3, the levels of IGF-1 in the Nef-containing groups were lower than those in the PLGA/BGP group. Conversely, at 7 days, IGF-1 expression was higher in the Nef-containing groups than in the PLGA/BGP group. IGF-1R expression in the PLGA/BGP/25N and PLGA/BGP/50N groups was obviously upregulated compared with that in the PLGA group at days 3 and 7. The expression level of the PLGA/BGP group was higher than that of the PLGA group only at 3 days and was the same as that of the PLGA group at 7 days.
The expression of CatK was shown in Fig.8. CatK is a cysteine protease with the highest expression and the strongest bone lytic activity in osteoclasts and is a key enzyme in the process of bone resorption. In this study, the expression level of CatK in RANKL‐treated osteoclasts was significantly downregulated in the PLGA/BGP/25N and PLGA/BGP/50N groups compared with the PLGA group at both 3 and 7 days.
3.2.7 Protein Expression by Western Blotting
In this study, MC-3T3-E1 and RANKL‐induced osteoclasts cultured on composite membranes for 7 days were collected. For MC-3T3-E1 cells, IGF-1-mediated osteogenic differentiation pathway proteins were detected. Western blotting (Fig.7) results indicated that, compared with the PLGA and PLGA/BGP groups, the amount of IGF-1-mediated osteogenic differentiation pathway proteins (IGF-1R, p-AKT and p-mTOR) produced in the Nef-containing groups increased, and the expression levels in PLGA/BGP/N25 were the highest.
For RANKL‐induced osteoclasts, NF-κB pathway-related pathway proteins and the osteoclast differentiation-related transcription factor NFATC1 were detected (Fig.9). Compared with the PLGA and PLGA/BGP groups, the phosphorylation of NF-κB pathway-related pathway proteins (p-p65 and p-IκBα) were reduced in the Nef-containing groups decreased. The expression of NFATC1 was also significantly inhibited (p<0.05).
3.2.8 Repair Effect of Skull Injury
The changes in skull defects over time observed by micro-CT are shown in Fig.10. The defects in each group were gradually covered by new bone tissue. The PLGA/BGP/Nef group had the highest regeneration coverage at all time points (p<0.05), exceeding 80% coverage at day 28. The regeneration coverage rate was approximately the same in the PLGA and PLGA/BGP groups, reaching approximately 50% after 56 days, while that of the control group was less than 20%. The BVF at the defect site of PLGA/BGP/Nef was also significantly higher than that of the other groups at each time point (p<0.05), up to approximately 70% at day 56. The BVFs of the PLGA/BGP group, PLGA group, and control group decreased in sequence. However, the difference between the PLGA/BGP group and PLGA group was not significant.
H&E and Masson staining were used to highlight typical features in tissue sections. Fig.11(A) shows the results of H&E staining. Connective tissue was formed along the occupation of material. The occupying part of the new bone tissue grew and gradually covered the defect, while the defect did not form a complete osseous connection in the control group. The implants act as a “bone bridge”. Especially in the PLGA/BGP/Nef group, the new bone formed thicker calli, and the repair effect was significantly better than that in the other groups. Fig.11(B) shows the results of Masson staining. Most of the defect sites in each group had been covered with newborn tissue. Scattered mature bone tissue stained red can also be observed inside the blue new bone. The thickness of new bone in the PLGA/BGP/Nef group was thicker than that in the other groups. There was also more red staining in the new bone site of this group. The results further confirmed that PLGA/BGP/Nef could enhance the effect of bone defect repair.